Jeff Ziegler

The European Political Science Society is now accepting paper & panel proposals for its annual conference!

📢 Call for Papers: EPSS 2026 – Belfast

🗓️ June 18–20, 2026

📍 ICC Belfast

📬 Deadline: Nov 7, 2025

🧵

This looks interesting.

"Bounding causal effects in survey experiments with noncompliance or inattention"

www.dropbox.com/scl/fi/wvzck...

Will anyone review this paper? Screening, sorting, and the feedback cycles that imperil peer review Scholarly publishing relies on peer review to identify the best science. Yet finding willing and qualified reviewers to evaluate manuscripts has become an increasingly challenging task, possibly even ...

1. Kevin Gross and I just posted a new science-of-science preprint.

This one explores the looming peer review crisis. As many of you know, it's becoming significantly more difficult for journal editors to find scholars willing to serve as peer reviewers for submitted manuscripts.

Open science contributes to more rigorous and impactful science. However, little attention is often paid to the benefits that it provides to its users, such as project management and programming skills. Although not its primary objective, these skills may be considered additional benefits supporting the transition to open science practices. Drawing from international perspectives, this chapter will discuss skills that students develop explicitly or implicitly by engaging in open science and their benefits to careers within and outside of academia. It will also showcase examples of how engagement in open science is considered in (academic) hiring decisions, including ongoing debates and areas for (structural) improvement. Ultimately, this chapter will inform educators about the most important open science skills and provide insights into how to strategically build a resume to present open science-related skills convincingly to improve employability, including practical tips and resources.

#OpenScience principles help students gain lots of useful skills that are an advantage on the job market in and beyond academia. Read our latest book chapter where we provide tips for educators to bolster employability through open science: doi.org/10.31219/osf...

Learn Stan with brms, Part I | A. Solomon Kurz y ~ 1

New #rstats blog up!

solomonkurz.netlify.app/blog/2025-07...

This is the first in a brief series where we use {brms} to learn {Stan} code.

Many thanks to @fusaroli.bsky.social and @stephenjwild.bsky.social for their helpful reviews.

@kaylynjschiff.bsky.social @dschiff.bsky.social

Thanks for the encouragement, and this is thankfully not an area of research that this applies to 😅

For the record, I hate thinking like this and don’t think it’s necessarily good for science for us to table work just because it takes a while to find a home

I imagine this is largely personal re: time and resources, but I’m really curious (and impressed) how others know when to prioritize work that will have better payoffs.

After how many failed submissions (and after incorporating reviewers’ feedback as genuinely as possible along the way) do you finally give up on a project? Asking for a friend…

this post uses newly released data from the 2020-2022-2024 CES panel survey, which includes interviews with ~6k Americans we interviewed (and re-interviewed) over the past three election cycles. download it and dig in! dataverse.harvard.edu/dataset.xhtm...

New NEH-supported tutorial on running LLMs locally with ollama! Your laptop is more powerful than you think. Save money, privacy, and energy.

aiforhumanists.com/tutorials/

Emergent LLM behaviors are observationally equivalent to data leakage Ashery et al. recently argue that large language models (LLMs), when paired to play a classic "naming game," spontaneously develop linguistic conventions reminiscent of human social norms. Here, we sh...

Our conclusion: Apparently "spontaneous" or "emergent" dynamics are observationally equivalent to data leakage.

That is, what looks like emergent behavior may be LLMs drawing on their pretraining data.

Read the full paper: arxiv.org/abs/2505.23796

Incorporating informative Bayesian priors offers the ability to learn more from experiments. @leoiacovone.bsky.social , @economeager.bsky.social & @dmckenzie.bsky.social demonstrate how to do this in the context of a Colombian exporting experiment. buff.ly/CwdGmKE

Such a cool book www.economics.utoronto.ca/osborne/mpe/...

The Experimental Research section of the American Political Science Association (@experimentsapsa.bsky.social) invites research proposals from post-prospectus PhD students for its Experimental Research Early-Career Fellowship program. Share this post! @apsa.bsky.social, @poscresearch.bsky.social

Writing some paragraphs about odds ratio and, more generally, different scales in nonlinear models.

Any favorite articles on odds ratio?>

You’re Not Blocked: You’re Just Trying to Write the Whole Thing at Once Break the Work Before It Breaks You

😶‍🌫️ You’re not blocked. You’re just trying to write the whole thing at once.

🧵 about writing, Einstein & big projects.

Prompted by great Q @dirckdekleer.bsky.social: How do I break down a big project into writing tasks that feel manageable?

🔗 catherineeunicedevries.substack.com/p/youre-not-...

We sometimes need to deviate from a preregistration to increase the validity of our inferences. Not all deviations reduce the severity of our tests - but they often do. Be transparent about all deviations, and evaluate the consequences for test severity.

online.ucpress.edu/collabra/art...

The quiet collapse of surveys: fewer humans (and more AI agents) are answering survey questions I show data on two trends undermining surveys: the collapse of human response rates and the increase of AI agents. I'll also discuss downstream implications and propose some possible solutions.

I wrote a new Substack! "The quiet collapse of surveys: fewer humans (and more AI agents) are answering survey questions" laurenleek.substack.com/p/the-quiet-...

The quiet collapse of surveys: fewer humans (and more AI agents) are answering survey questions I show data on two trends undermining surveys: the collapse of human response rates and the increase of AI agents. I'll also discuss downstream implications and propose some possible solutions.

Fascinating substack post by @laurenleek.eu on the problem of declining humans and increasing AI agents in survey responses. This is going to become a huge issue in survey research!
open.substack.com/pub/laurenle...

From Weingast's essay "Caltech Rules"

"With rare exceptions, papers do not write themselves. Transforming a good idea into a good paper is a difficult process. A clear understanding of what each part of your paper must accomplish is essential to this process."

Link to essay weingast.people.stan...

#RDDers: check it out.

rd2d: Causal Inference in Boundary Discontinuity Designs

"This article introduces the R package rd2d, which implements and extends the...results developed in Cattaneo et al. (2025) for boundary discontinuity designs [like geographic RDDs]"

arxiv.org/abs/2505.07989

TWFE Event_study plot showing sign reversal

I created a shiny web tool to play around with OL/TWFE so you can teach how this stuff can get so screwy. You can mess with temporal and cohort heterogeneity, treatment timing, whether you have any controls, etc. Share your worst plots! #econsky cannoncloud.shinyapps.io/TWFE_OLS_Pla...

Zack Elkins (@zachelkins.bsky.social) compiled a wonderful collection of perspectives on academic writing (esp. as a political scientists).

I use it often.

His page: sites.google.com/site/zachelk...

I made a backup here as well, so nothing is lost: github.com/carlislerain...

🧵 What kind of POPE do AMERICANS want?📸

Before the Conclave, @sophiemainz.bsky.social and I asked 1,500 U.S. adults to choose among hypothetical papal candidates.

Results: Americans prefer a liberal and *non-political* Pope. The latter might be challenging for Leo XIV.

Results ⬇️

Polisky Datasky

On the verge of declaring defeat with chatgpt in my asynchronous online dataviz class. Something changed this semester compared to past ones and SO MANY assignments are essentially 100% LLM output.

If you're a grad student or an undergrad interested in research I need to you listen to me very carefully.

You cannot learn to write good research papers if you do not read good research papers.

Stop asking LLMs to summarize papers for you.

Ten principles for reliable, efficient, and adaptable coding in psychology and cognitive neuroscience - Communications Psychology Programming is essential for modern research in neuroscience and psychology, but it can quickly become a source of frustration and error. This Primer introduces ten practical principles guiding resear...

Many processes from industry are not appropriate or overkill for an academic setting, but we can and should be doing better than we currently are.

Our paper tries to give practical advice on how you can improve your scientific workflow, without being a coding guru.

www.nature.com/articles/s44...